Why Is DeepSeek Raising Another $7 Billion Just Months After Its First Round? Inside the $70 Billion Valuation Talks

If Round 2 closes, DeepSeek will have raised over $14 billion in under five months. The Chinese AI lab behind open-weight R1 and V4 models restarted talks on August 4–5 for a second external round targeting roughly 50 billion yuan (~$7 billion) at a pre-money valuation of about 500 billion yuan (~$70 billion) — up 43% from its June post-money figure. This article walks through the timeline, cap-table voting rights, 140–150x P/S math, peer comparison, and STAR Market IPO backdrop. Every Round 2 figure below comes from anonymous dealmakers cited by Chinese financial media, not an official DeepSeek statement.

SECTION 01 Four signals to watch before treating the Round 2 numbers as settled

• Round 2 terms are still in negotiation, not closed: The target raise of 50 billion yuan, pre-money valuation of ~500 billion yuan, and late-August signing window all come from Caijing magazine citing unnamed dealmakers. DeepSeek has not publicly confirmed any of it, and terms could still shift.
• The valuation-to-revenue gap is extreme: Reported annualized revenue (ARR) of $400–500 million implies a ~140–150x price-to-sales ratio — well above dealmaker estimates of ~65x for OpenAI and ~21x for Anthropic. Pricing looks more like an options bet than a cash-flow valuation.
• Most outside investors have no vote: In Round 1, most capital flowed through a limited partnership controlled by founder Liang Wenfeng — no voting rights, five-year lock-up. Only China's National AI Industry Investment Fund invested directly, with voting rights and no lock-up. Governance and state-influence questions remain sensitive.
• Talks already paused once over a leaked transcript: On July 25–26, Round 2 stalled after Liang reportedly grew unhappy that remarks from closed-door investor meetings had circulated online. Both sides restarted on August 4–5 and reportedly want to keep this round low-profile.

Timeline:

• April 2026: A corporate filing shows registered capital increased and Liang Wenfeng raised his direct stake from 1% to 34%; combined with an entity he controls, total control reached roughly 84.29%. That same month, DeepSeek opened its first external funding round and previewed the V4 model series. See DeepSeek V4 GA release breakdown.
• June 2026: Round 1 closed at roughly 50 billion yuan (~$7.4 billion), post-money valuation above 350 billion yuan (reported range $52–59 billion across sources) — the largest first-round raise in Chinese AI history. Liang personally contributed 20 billion yuan; Tencent 10 billion yuan, CATL 5 billion yuan, with JD.com, NetEase, IDG Capital, and China's National AI Industry Investment Fund also participating.
• July 14–17, 2026: Multiple outlets reported DeepSeek began preparing a STAR Market IPO and was in talks for Round 2 at a pre-money valuation of roughly $71 billion (~480 billion yuan). DeepSeek's ARR — reportedly $400–500 million, mostly from API token usage — became public for the first time.
• July 25–26, 2026: Round 2 talks abruptly paused.
• August 4–5, 2026: Dealmakers cited by Caijing said the round had restarted, still targeting 50 billion yuan at ~500 billion yuan pre-money (~$70 billion), with signing expected in late August. Product-side context: DeepSeek V4 Flash official benchmarks.

Every figure above about Round 2 — the amount, the valuation, the timeline — comes from anonymous dealmakers cited by Chinese financial media, not from an official DeepSeek statement. Terms could still shift before signing.

SECTION 02 Round 1 vs. Round 2: the numbers at a glance

The tables below summarize reported funding pace and valuation basis for quick cross-checking. Dollar conversions vary by source; yuan figures follow Chinese media reporting.

SECTION 03 What the $70 billion valuation is actually pricing: compute, votes, and multiples

1. The real bill is compute, not headlines. Shortly after closing Round 1, DeepSeek said it would double headcount across data-center and AI-agent teams, and Reuters reported it was hiring chip-design engineers to develop its own AI inference chips. Industry analysts estimate that for every 10 billion yuan DeepSeek raises, roughly 7 billion yuan goes straight into compute-related spending — chips, data centers, bandwidth, liquid cooling. The fundraising cadence is a race to keep pace with its own compute buildout, not a valuation bragging exercise.

2. Most investors do not get a vote. In Round 1, most outside capital flowed through a limited partnership controlled by Liang Wenfeng — no voting rights, five-year lock-up. The one exception: China's National AI Industry Investment Fund, which invested directly and received both voting rights and no lock-up. This structure keeps Liang's control near 84% and has drawn scrutiny from outlets like Forbes about governance and state influence.

3. A 148x price-to-sales ratio is either a bet on the future — or a red flag. At a $70 billion pre-money valuation against $400–500 million in ARR, DeepSeek's implied P/S sits around 140–150x, dwarfing OpenAI's ~65x and Anthropic's ~21x. One dealmaker's assessment, translated from Chinese coverage, is blunt: "Pricing a foundation-model company is fundamentally an options bet, not a cash-flow valuation." Investors are pricing the chance that DeepSeek becomes infrastructure-level in China's compute ecosystem and enterprise agent market.

SECTION 04 How DeepSeek stacks up against Moonshot, Zhipu, and MiniMax

DeepSeek and Moonshot — the two still-private frontrunners — both carry P/S multiples around 140–150x, well above what listed Zhipu and MiniMax trade at in the secondary market. Private-market investors are paying a steeper premium for labs that have not yet faced public-market scrutiny. Open-weight competition context: Kimi K3 open-weight release breakdown.

SECTION 05 A leaked transcript, voting-rights scrutiny, and bubble warnings

1. A leaked closed-door transcript stalled the deal. The immediate trigger for the July pause was reportedly Liang Wenfeng's frustration that remarks from first-round investor meetings had spread online — a reminder that as DeepSeek's investor base grows, keeping a low profile is getting harder.
2. The voting-rights structure is drawing outside scrutiny. Most external investors have no vote and a five-year lock-up, while only the state-backed National AI Industry Investment Fund gets direct voting rights with no lock-up. Forbes and CIW have flagged governance questions; DeepSeek has not publicly addressed them.
3. The valuation-to-revenue gap remains unresolved. A 140–150x P/S ratio is extreme by any industry's standard — even top-tier, high-growth SaaS companies typically trade at 30–50x. Whether this valuation holds depends on whether DeepSeek can convert its technical lead into scaled enterprise revenue after a STAR Market listing — something the market has not yet tested.

Every detail above about deal size, valuation, and ownership structure comes from anonymous-sourced reporting (Caijing, Reuters, Bloomberg, Forbes, among others). DeepSeek has not officially confirmed Round 2 terms.

SECTION 06 STAR Market rule changes, China's compute push, and the global AI funding race

• China rewrote listing rules for unprofitable AI companies. On June 17, 2026, the Shanghai Stock Exchange announced at the Lujiazui Forum that it was expanding its "fifth listing standard" on the STAR Market to cover AI companies — no profit or significant revenue required, as long as technology is strong enough. That rule change is the regulatory backdrop for DeepSeek's planned late-2026 IPO filing (targeting a 2027 listing).
• DeepSeek abandoned a five-year "no fundraising, no IPO, no commercialization" policy. For years, DeepSeek was funded entirely by founder Liang Wenfeng's quant fund, High-Flyer, and refused outside capital. That policy ended with Round 1 in June 2026 — a symbolic moment as rivals Zhipu and MiniMax have listed in Hong Kong and Moonshot keeps raising at an accelerating pace.
• This is part of a global re-pricing of frontier AI labs. OpenAI was reportedly valued at $300 billion in 2025, and Anthropic's valuation reportedly surpassed OpenAI's by June 2026. Investors paying a steep premium for a globally competitive Chinese lab is, in part, a bet that Chinese model companies can keep pace technically while serving a market as large as any economy's.
• Compute self-reliance is the subtext. Reports that DeepSeek is developing its own AI inference chips and building out its own data centers mirror a broader trend among leading Chinese AI labs to pair frontier model development with domestic compute and chip strategies — which also explains why a company with comparatively modest revenue still needs to raise capital this fast and this often.

SECTION 07 Six-step checklist after reading the DeepSeek funding headlines

1. Separate "reported" from "closed": Round 2 target amount, pre-money valuation, and signing window are not officially confirmed. Cite figures as "per dealmakers cited by media."
2. Check whether the valuation is pre-money or post-money: Round 1 is commonly cited as "post-money above 350B yuan"; Round 2 reporting uses "pre-money ~500B yuan." Mixing bases distorts the growth math.
3. Stress-test with ARR and P/S: At $400–500 million ARR, implied multiples sit around 140–150x — compare against OpenAI, Anthropic, and listed Chinese peers.
4. Read the cap table and voting rights: Most LP investors have no vote and a five-year lock-up vs. the National AI Industry Investment Fund's direct stake. Governance and exit paths depend on this structure.
5. Map fundraising pace to compute capex: Analysts estimate ~70% of each raise goes to chips, data centers, bandwidth, and liquid cooling. Funding cadence reflects a capex race, not just valuation momentum.
6. Track STAR Market timeline alongside product releases: IPO prep (targeting late-2026 filing, 2027 listing) runs parallel to V4-series iteration. Product capability still drives API and agent production value.

SECTION 08 Citeable data points and primary sources

• Round 2 target: ~50 billion yuan raise; ~500 billion yuan pre-money valuation; signing planned late August 2026 (Caijing citing dealmakers)
• Round 1 closed: ~50 billion yuan; post-money above 350 billion yuan; Tencent ~10 billion yuan, CATL ~5 billion yuan among participants
• Valuation step-up: Round 2 pre-money ~+43% vs. Round 1 post-money; media describe roughly 7x growth from ~$10B-class levels in April to ~$70B pre-money in under four months
• ARR / P/S: ~$400–500 million ARR; implied P/S ~140–150x
• Control: Liang Wenfeng and related entities control roughly 84.29% of equity (corporate filing coverage)

Third-party reporting links — verify latest developments before citing:

Sina Finance — DeepSeek restarts second funding round at 500B yuan pre-money

Gate News — DeepSeek reopens second funding round

ChainCatcher — DeepSeek restarts financing coverage

Forbes — DeepSeek first-round governance coverage (search: "DeepSeek Just Raised $7.4 Billion. Here's The Catch.")

Funding headlines aside, production teams still face API bill volatility, agent integration cost, and whether local compute stays under control. Pure cloud token pricing swings with peak billing and cache-hit rates; virtualized cloud Macs add compatibility and performance overhead. For environments that need zero-loss native compute, stable iOS CI/CD, and 7×24 agent automation, VPSNIX cloud physical nodes are usually the better production substrate — 100% Apple hardware, full root access, zero hypervisor tax, elastic day/week/month billing. Pair with Mac mini M4 rental pricing analysis and the pricing page for a two-tier stack.

SECTION 09 FAQ